Torres‑Haro et al. Microb Cell Fact (2021) 20:175 https://doi.org/10.1186/s12934‑021‑01664‑6

REVIEW

Metabolic engineering for high yield synthesis of astaxanthin in Xanthophyllomyces dendrorhousAlejandro Torres‑Haro , Jorge Verdín , Manuel R. Kirchmayr and Melchor Arellano‑Plaza*

Abstract

Astaxanthin is a carotenoid with a number of assets useful for the food, cosmetic and pharmaceutical industries. Nowadays, it is mainly produced by chemical synthesis. However, the process leads to an enantiomeric mixture where the biologically assimilable forms (3R, 3′R or 3S, 3′S) are a minority. Microbial production of (3R, 3′R) astaxanthin by Xanthophyllomyces dendrorhous is an appealing alternative due to its fast growth rate and easy large‑scale production. In order to increase X. dendrorhous astaxanthin yields, random mutant strains able to produce from 6 to 10 mg/g dry mass have been generated; nevertheless, they often are unstable. On the other hand, site‑directed mutant strains have also been obtained, but they increase only the yield of non‑astaxanthin carotenoids. In this review, we insight‑fully analyze the metabolic carbon flow converging in astaxanthin biosynthesis and, by integrating the biological features of X. dendrorhous with available metabolic, genomic, transcriptomic, and proteomic data, as well as the knowledge gained with random and site‑directed mutants that lead to increased carotenoids yield, we propose new metabolic engineering targets to increase astaxanthin biosynthesis.

Keywords: Astaxanthin, Carotenoids, Biological pigments, Metabolic engineering, Omics‑based engineering, Xanthophyllomyces dendrorhous

© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom‑mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

IntroductionCarotenoids have gained biotechnological importance due to their benefits to human health. They are power-ful antioxidative and anti-inflammatory agents, as well as efficient UV radiation protectants. In addition, they are precursors of retinol (Vitamin A) and other derivatives that participate in cell regeneration. Therefore, carot-enoids are interesting for food, cosmetic and pharma-ceutical industries [1–5]. Among the carotenoids with the highest biotechnological potential are phytoene, lyco-pene, β-carotene, zeaxanthin and astaxanthin [1, 6, 7].

Astaxanthin (3,3′-dihydroxy-β, β′-carotene-4,4′-dione) is probably one of the most interesting and valuable carotenoids. Its antioxidant activity has been reported to be 10 and 100 times higher than those of β-carotene and α-tocopherol, respectively [8–10]. It is estimated that by the end of 2020 there will be a turnover in carotenoids sales of around $1.85  billion USD. The sales of astax-anthin represents 29% of the total with a 2.3% annual increase [11]. Total astaxanthin sales by 2025 are esti-mated to be $2.57  billion [12]. Synthetic astaxanthin price is around 2500 USD/kg, but the price increases to about $100,000 USD/kg for natural and high purity asta-xanthin [13–15].

Value differences between natural and synthetic asta-xanthin arise from the composition of the enantiomeric mixture each source produces, which is also related to their biological activity. There are three possible

Open Access

Microbial Cell Factories

*Correspondence: marellano@ciatej.mxBiotecnología Industrial, Centro de Investigación y Asistencia en Tecnología y Diseño del Estado de Jalisco, A.C. Camino Arenero 1227, Col. El Bajío del Arenal, 45019 Zapopan, Jalisco, Mexico

Page 2 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

astaxanthin stereoisomers: (3R, 3′R), (3R, 3′S) and (3S, 3′S). Natural astaxanthin is predominantly (3S, 3′S) or (3R, 3′R), while synthetic astaxanthin contains a mixture of all three possible enantiomeric forms, (3R, 3′R), (3R, 3′S), and (3S, 3′S), in a 1:2:1 ratio, respectively [16–18]. Because of the enantiomeric mixture, synthetic astaxan-thin shows relatively low biological activity [17–19]. On the other hand, the enantiomeric form of natural astax-anthin may vary from one organism to another; (3R, 3′R)

is the main astaxanthin stereoisomer produced by Xan-thophyllomyces dendrorhous (Fig. 1A) [17, 20], while the (3S, 3′S) form is mainly produced by Haematococcus plu-vialis [17, 21].

Bernhart et  al. [16] analyzed the natural accumula-tion of different astaxanthin stereoisomers in Meg-anyctiphanes norvegica, and two different species of salmonids: Atlantic salmon (Salmo salar) and Pacific salmon (Oncorhynchus tshawytscha). They observed

Fig. 1 A Chemical structure of the astaxanthin (3R, 3R) isomer produced by X. dendrorhous. B General scheme of astaxanthin biosynthesis pathway in X. dendrorhous. Genes, enzymes, and substrates that participate in the metabolic pathway are shown in red, blue, and black, respectively

Page 3 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

that (3S, 3′S)-astaxanthin accumulated more (77–85%) than (3R, 3′R) stereoisomers (11–18%). However, Foos et  al. [17] reported a similar absorption of both (3R, 3′R) and (3S, 3′S) stereoisomers (90% and 96%, respectively) in rainbow (Oncorhynchus mykiss) and sea (Salmo trutta) trout. Similarly, a study in humans showed that the absorption of both (3R, 3′R and 3S, 3′S) stereoisomers was also similar [19].

In terms of bioactivity, non-esterified (3S, 3′S and 3R, 3′R)-astaxanthin applied to lymphocytes and peritoneal exudates of mice [22], promote cell proliferation, killer cell cytotoxic and phagocytic activity. Nevertheless, non-esterified (3S, 3′S)-stereoisomer showed a slightly higher immunoregulatory effect than the other one. (3R, 3′R)-astaxanthin shows specific activities as lipop-eroxidation reducing agent in rainbow trout [23], anti-oxidant capable of delaying aging [24] and, in chickens, increases immunoglobulin and splenocyte proliferation [25].

Xanthophyllomyces dendrorhous natively produces around 97% of non-esterified (3R, 3′R)-astaxanthin [14, 26]. In contrast, H. pluvialis produces (3S, 3′S)-astaxan-thin in high yields, but it is esterified by palmitic, stearic or linoleic acids [13, 14, 17, 18], which requires hydrolysis by lipases before it can be assimilated. The assimilation of astaxanthin esters may be variable due to the complex-ity of the molecule and the organism assimilation capac-ity [16, 17, 19]. Storebakken et al. [18] reported a lower absorption of astaxanthin dipalmitate esters (47%) than free astaxanthin (64%) in Atlantic salmon.

Nowadays, most astaxanthin is produced by chemi-cal synthesis; nevertheless, microbial sources are well known. As can be inferred from previous paragraphs, the main producers of this pigment are the yeast X. den-drorhous and the microalgae H. pluvialis [1, 12, 26–30]. Nevertheless, microbial production of astaxanthin is less competitive than the synthetic one because of difficulties to obtain enough biomass and its low pigment yield, thus overproducing astaxanthin from appropriate microbial sources is a current biotechnological challenge [1, 12, 31]. So far, bioengineering approaches that include cul-ture media and bioreactor optimization [30], as well as construction of improved overproducing strains [12, 31, 32], have been assayed. Metabolic engineering of Saccha-romyces cerevisiae, Candida utilis, Pichia pastoris and X. dendrorhous have led to increased amounts of phytoene, lycopene or β-carotene, but they have failed to increase astaxanthin biosynthesis in relevant levels [1, 30, 33–35].

Here, we envisage new metabolic engineering targets to generate X. dendrorhous strains able to shift the car-bon metabolic flux towards astaxanthin biosynthesis. To that end, random and site-directed genetic modifications made to X. dendrorhous and other yeasts were analyzed

from their consequences in gene expression to the impact in astaxanthin yield.

Xanthophyllomyces dendrorhous biologyThe yeast X. dendrorhous (also known as Phaffia rho-dozyma in its teleomorphic state) is a basidiomycete that exceptionally produces xanthophylls in response to lipoperoxidation caused by environmental stress [36, 37]. In addition to pigments, X. dendrorhous genome encodes a vast battery of anti-oxidative enzymes such as catalases and superoxide dismutases [38]. X. den-drorhous life cycle is homothallic and characterized by the mating of a mother cell with a bud (paedogamy) on polyol-rich media. After mating, a non-septated basid-ium (holobasidium) is formed with mononuclear basidi-ospores arising on its apex [37, 38]. These basidiospores lead to diploid vegetative cells [39–41], although haploid strains have been characterized [42]. Unlike basidiomy-cetous yeasts, X. dendrorhous does not show a transition from unicellular to filamentous growth during the sexual phase.

The elucidation of the genome sequence of several X. dendrorhous strains (19.5 megabases, and 6385 protein coding genes in CBS 6938 strain) showed an atypical high rate of introns per gene (7.4–7.5) [38, 42]. Additional genomic analysis also revealed the existence of two main acetyl-CoA derived biosynthetic pathways, terpenoid and fatty acid biosynthesis [42]. Terpenoids in X. den-drorhous, like in any other fungus, are synthesized via the mevalonate pathway and typically ends with ergosterol synthesis. However, as shown in the next sections, the mevalonate pathway in X. dendrorhous bifurcates, lead-ing to the carotenoids and sterols pathways. Unlike other carotenogenic fungi, genes involved in sterols and carot-enoids biosynthesis in X. dendrorhous are not organized in clusters.

Biosynthesis of astaxanthin in X. dendrorhousAstaxanthin derives from the isoprene biosynthesis pathway (Fig.  1B). Acetyl-CoA produced after glycoly-sis is introduced into the mevalonate pathway to yield isopentenyl pyrophosphate (IPP) [17, 20]. IPP can be isomerized in dimethylallyl pyrophosphate (DMAPP) by bifunctional IPP isomerase (idi). Later, geranyl pyrophosphate (GPP) and farnesyl pyrophosphate (FPP) are synthesized by the action of FPP synthase (ERG20). Afterwards, the ligation of IPP with FPP, mediated by GGPP synthase (crtE), leads to geranylge-ranyl pyrophosphate (GGPP). The condensation of two GGPP molecules carried out by phytoene-β-carotene synthase (crtYB) leads to phytoene. Subsequently, four dehydrogenations on phytoene are performed by phy-toene desaturase (crtI) to synthesize lycopene. Two

Page 4 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

cyclization steps catalyzed by phytoene-β-carotene syn-thase (crtYB) are needed to finally produce γ-carotene and β-carotene. Astaxanthin formation involves the oxidation of β-carotene by astaxanthin synthase (crtS) and cytochrome P450 reductase (crtR) [43–47]. After astaxanthin biosynthesis, lipoproteins transport the molecule into biological membranes for its storage and cell protection [6].

Glyceraldehyde 3-phosphate and pyruvate also contrib-ute to the mevalonate pathway since the metabolic flux of these compounds produce acetyl-CoA [48–50]. Nev-ertheless, pyruvate and acetyl-CoA are also amino acid and fatty acids precursors, as well as substrates of the tri-carboxylic acids cycle (Fig. 1B) [51]. Although these sub-strates cannot be exclusively dedicated to the astaxanthin biosynthesis, it is possible to stress the yeast, either mod-ulating the fermentation conditions or adding precursors and cofactors, to improve astaxanthin yields [30, 49].

Xanthophyllomyces dendrorhous, despite its high capacity to accumulate astaxanthin, shows a number of limitations that hampers its synthesis [26, 52]. Within the mevalonate pathway, hydroxymethylglutaryl-CoA reduc-tase (HMG1), the main supplier of the pathway (Fig. 1B), is regulated by product and, therefore, it is a limiting step that hinders the carbon flux [53]. In addition, the accu-mulation of 3-hydroxy-3-methylglutaryl-CoA, substrate of the aforesaid reductase, is toxic to cells, which explains the low expression levels of genes involved in its synthe-sis. A second limiting step is the reaction catalyzed by mevalonate kinase (ERG12) whose expression is regu-lated by isoprenoid products, mainly geranyl pyrophos-phate and farnesyl pyrophosphate [54].

Within the astaxanthin biosynthesis pathway, phytoene-β-carotene synthase (crtYB) is also a bottleneck for the accumulation of astaxanthin due to its low expres-sion levels [33, 55–58]. Similarly, low expression levels of cytochrome P450 reductase (crtR), which participates in the catalysis of the last step of the astaxanthin biosyn-thesis, have been observed [51, 56, 59–61]. Moreover, the biosynthesis of carotenoids is inactivated when catabolite repression is induced. In X. dendrorhous, Mig1 has been identified as a catabolic repressor of carotenoids biosyn-thesis [62].

Table 1 shows the enzymes and encoding genes directly involved in the astaxanthin biosynthesis, and their sub-cellular localization within X. dendrorhous.

Biotechnological production of astaxanthin by X. dendrorhousXanthophyllomyces dendrorhous is the most widely used microorganism in industry for astaxanthin production since, using appropriate substrates, it is its main carot-enoid (85% of total, approximately) [26]. However, the low yield of astaxanthin produced by the yeast (which varies between 200 and 400 µg/g of dry biomass) and the high costs of culture media, have hampered its use as a large-scale source [20, 30, 46, 63]. Moreover, the chemi-cal synthesis of astaxanthin has a final yield of 8% in rela-tion to supplies employed in the process [64]. This value is higher than the biological production of astaxanthin in X. dendrorhous whose yield lays between 0.02 and 0.03% with respect to the substrate employed [1, 63, 65, 66].

In order to make microbial production more com-petitive, it is obviously necessary to increase the yield

Table 1 Genes/enzymes involved in the biosynthesis of astaxanthin by X. dendrorhous

MIS mitochondrial intermembrane space, ERM endoplasmic reticulum membrane, BM vacuole membrane, ICM integral component of membrane, MOM mitochondrial outer membrane, PM plasma membranea Reference [129]

Gene name Gene ID Enzyme name Subcellular localizationa References

ERG10 CED82454 Acetyl‑CoA acyltransferase Cytoplasm, MIS [42, 126, 129, 130]

ERG13 CED83016 Hydroxymethylglutaryl‑CoA synthase Nucleus [42, 126, 129, 130]

HMG1 CED85502 Hydroxymethylglutaryl‑CoA reductase Nucleus periphery, ERM [42, 126, 129, 130]

ERG12 CED82307 Mevalonate kinase Cytoplasm [42, 126, 129, 130]

ERG8 CDZ97430 Phosphomevalonate kinase Cytosol and nucleus [42, 126, 129, 130]

MVDD CED83492 Diphosphomevalonate decarboxylase Cytosol [42, 126, 129, 130]

idi CED82414 Isopentenyl pyrophosphate isomerase Cytoplasm and nucleus [42, 126, 129, 130]

ERG20 CDZ96456 Farnesyl pyrophosphate synthase ERM [42, 126, 129]

crtE CDZ97186 Geranylgeranyl pyrophosphate synthase [42, 126, 129, 130]

crtYB CED83449 Phytoene‑β‑carotene synthase [126, 129, 130]

crtI CED83513 Phytoene desaturase ICM [126, 129, 130]

crtS CED83940 Astaxanthin synthase [126, 129, 130]

crtR CDZ98161 Cytochrome P450 reductase MOM, ERM, PM [126, 129, 130]

Page 5 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

[12, 30, 31, 63]. Nutrimental and operational fermenta-tion conditions have been optimized to stimulate asta-xanthin biosynthesis in X. dendrorhous, among which can be accounted carbon source concentration (15.0 to 35.0 g/L), nitrogen (0.5 to 3.0 g/L), pH (5.5 to 6.0), tem-perature (18.0 to 22.0 °C), dissolved oxygen (above 20%) [29, 30, 65, 67–71], white light (177 μmol photon/m2 × s) [13, 72], and inoculum size (5–10%) [69]. Dissolved oxy-gen has been elucidated as one of the most important factors in X. dendrorhous fermentation. Wu et al. [73], by means of a transcriptional analysis, showed that supply of 25% dissolved oxygen leads to 2.5-fold increase of crtYB and crtI expression, and 1.5-fold increase of crtR and crtS expression. Regarding nutritional requirements, glucose and ammonium sulfate are the preferred carbon (C) and nitrogen (N) sources, which maximize astaxanthin pro-duction when they are employed in a high C/N molar ratio (usually between 40 and 76) [45, 46, 51]. In addition, glutamate and yeast/malt extracts increase the assimila-tion of C/N for astaxanthin biosynthesis [74].

Other strategies to increase astaxanthin yields in X. dendrorhous and other yeasts, even more promising than fermentation optimization, is the generation of random [75, 76] or site-directed [30, 33, 55, 77] overproducing mutant strains [8, 30, 47, 49, 75, 79].

Physical chemical mutagenesis of X. dendrorhous to increase astaxanthin biosynthesisRandom mutagenesis has been set up for the isolation of carotenoids-overproducing X. dendrorhous strains. These strategies consist in the exposure of yeast cells to physical and chemical mutagenic agents [8, 78–80] such as ultra-violet and gamma radiation, 1-methyl-3-nitro-1-nitros-oguanidine (NTG) [8, 45, 79–81] and/or ethyl methane sulfonate (EMS) [29, 45]. Because these strategies lead to a large number of mutant colonies, it is necessary to implement efficient selection methods to identify the best mutants [68, 79]. Astaxanthin and/or sterol biosyn-thesis inhibitors in the culture medium such as β-ionone, diphenylamine, ketoconazole, miconazole, 2-methyl-imidazole, clotrimazole, nicotine, nystatin, mevinolin, pyridine, and N,N-diethylamine facilitate the selection of mutants that overproduce carotenoids above wild-type levels because the mutant-enhanced metabolic pathway (carotenoids or biosynthetic precursors) can overcome the effect caused by the inhibitor [78, 79].

Gui-Li et al. [4] obtained X. dendrorhous Y119, a strain able to produce 6.4 mg astaxanthin/g of dry biomass, by exposing yeast cells to EMS. Jeong-Hwan et  al. [8] and Ang et al. [78] reported the use of NTG to generate JH1 and M34 strains, respectively. JH1 achieved an astaxan-thin production above 4 mg/g of dry biomass, while M34 only yielded 0.5 mg/g, i.e., 2 to 50 times more astaxanthin

than wild-type strains in batch culture. These results illustrate the great variability of the output of random mutagenesis approaches [17, 72].

In order to obtain genetically stable X. dendrorhous strains, Chun et  al. [82] performed a recombination method in which, in the first instance, native strains were subjected to chemical mutagenesis with NTG. The result-ing strains produced carotenoids above 1.6  mg/g of dry biomass after induction with antimycin A [46, 79] which, due to its chemical properties, alters the electron trans-port chain and, consequently, the mitochondria. These physiological affectations activate the cellular defense machinery, i.e., the biosynthesis of carotenoids [79]. After selection of the best mutants, they were genetically back-crossed by each other’s protoplast fusion to eliminate deleterious mutations. The resulting hybrids were mostly stable and capable of hyperproducing carotenoids in yields greater than 2 mg/g of dry biomass.

Random mutagenesis may have negative effects on cell physiology, viability, growth and metabolic capacity that cannot be overlooked [45, 79, 83]. These mutants are unstable and frequently lose their overproduction ability because of active cell repairing mechanisms [79, 84, 85]. Alternative strategies to develop stable mutants is the knocking-out and/or overexpression of genes directly or indirectly involved in the astaxanthin biosynthesis using site-specific molecular tools [30, 75, 86].

Metabolic engineering of X. dendrorhousThe increased availability of metabolic engineering tools has had a major impact on the development of X. den-drorhous strains that overproduce astaxanthin [12, 30–32, 49, 86–88]. These strains have been designed under the principle of maximizing the carbon flow from the basic substrates towards the astaxanthin biosynthesis route, minimizing the generation of unwanted byprod-ucts (Fig.  1B) [31, 49, 51, 88]. Shifting the carbon flux towards astaxanthin biosynthesis has been achieved by spotting key enzymes and bottlenecks of the metabolic pathway [31, 33, 51, 74], optimizing the expression levels of relevant genes, and thereby increasing the storage of astaxanthin in cells [33, 49, 51, 74, 86].

Besides the availability of X. dendrorhous genome sequence [42], efficient integrative vectors that contain positive selection markers such as nptII (G418) or hph (hygromycin), have been developed (Fig.  2). Those vec-tors have been used to knock-out or overexpress genes under the control of strong constitutive promoters. Recently, genetic engineering strategies have been imple-mented in X. dendrorhous to increase isoprenoids accu-mulation and consequent carotenoids biosynthesis [12, 30–32, 49, 86–88]. Moreover, Cre/lox-P systems have been used to remove selection markers after integration

Page 6 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

in X. dendrorhous genome [59]. Those molecular tools, in addition to the characteristic genetic tractability of X. dendrorhous, have stimulated an abundant metabolic engineering literature [12, 30, 31, 49].

In the next sections, intervention of the crt gene fam-ily that encodes enzymes of the carotenoids biosynthesis pathway is discussed.

Overexpression/inactivation of the crtYB geneOne of the main targets of metabolic engineering has been crtYB, a gene that encodes for phytoene-β-carotene synthase (Table  1), the bifunctional enzyme with phy-toene cyclase and β-carotene synthase activities respon-sible for the phytoene and β-carotene synthesis [54, 56, 58, 89].

Phytoene is the first carotenoid of the astaxanthin biosynthesis pathway (Fig.  1B). This carotenoid has a growing market as a skin protector and food supple-ment [90]. X. dendrorhous is a potential natural source of this compound (Table 2, Fig. 1B). Pollmann et al. [58] introduced genetic modifications to X. dendrorhous to obtain high yields of phytoene. To achieve this goal, crtI was knocked down to block the carbon flow and

allow phytoene accumulation (Fig.  1B). The phytoene biosynthesis was further improved by integrating 4 copies of HMG1 (encoding hydroxymethylglutaryl-CoA reductase). In addition, 16 copies of crtYB and 4 copies of crtE were introduced (Fig. 2B). The resulting strain was able to produce 7.5 mg phytoene/g biomass. In a continuous culture at a small scale, a greater effi-ciency was achieved with phytoene yields of more than 10 mg/g biomass. This result is higher than in any other reported organism [58], which demonstrates the poten-tial of X. dendrorhous for carotenoids biosynthesis [30].

Exploring the consequences of knocking crtYB out with a nptII-containing integrative cassette, Visser et al. [33] isolated white colonies unable to synthesize carot-enoids, which demonstrated the importance of crtYB. This result confirmed the work of Verdoes et  al. [55], who found that crtYB is a limiting step in the biosyn-thesis of carotenoids and astaxanthin [43–45, 47, 49].

Realizing the importance of phytoene-β-carotene synthase (crtYB), and keeping active the basal crtI function, multiple copies of crtYB were introduced into PR-1-104, a X. dendrorhous random mutant that already overproduced β-carotene (1 mg/g dry biomass) [89]. Overexpression of phytoene-β-carotene synthase

Fig. 2 Metabolic engineering of X. dendrorhous for overproduction of carotenoids. A Representation of knock‑in and knock‑out strategies for genetic engineering in yeast. B Overexpression of HMG1, crtE and crtYB for high production of phytoene [58]. C Genetic engineering of X. dendrorhous for high yield astaxanthin production. Integration of multiple copies of crtYB and crtS to increase the production of β‑carotene and its conversion to astaxanthin, respectively [33, 89]. D Metabolic engineering and conventional mutagenesis for high yield astaxanthin production [66]. crtYB and crtS genes were integrated using geneticin (G418) and hygromycin (hph) as selection markers. Afterwards, a random chemical mutagenesis was carried out using 200 µg/mL NTG. Chemical mutagenesis was repeated several rounds until the maximal astaxanthin overproducer was obtained. The best strain obtained, AXJ‑20/crtYB, was able to yield above 9.7 mg/g of dry biomass

Page 7 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

in such genetic background led to 1.5 times higher β-carotene accumulation than the parental strain.

In an independent work, Ledetzky et  al. [56] inte-grated 2 and 3 copies of crtYB in different strains and, in addition to an increase of canonical carotenoids, they detected the biosynthesis of a new carotenoid never described before, 3-HO-4-keto-7′,8′-dihydro-β-carotene. Both strains were able to produce more than 0.7 mg carotenoids/g of dry biomass, of which 40% was astaxanthin. The biosynthesis of new carotenoids was rationalized by the fact that phytoene synthase/lyco-pene cyclase can perform cyclizations in carotenoids. This enzyme has been shown to take phytoene and neurosporene as substrates to catalyze the sequential biosynthesis of β-zeacarotene, 7,8-dihydro-β-carotene and, with the participation of crtR and crtS [43, 44, 47, 75], 3-HO-4-keto-7′,8′-dihydro-β-carotene (Fig.  2B) [56]. crtS encodes for astaxanthin synthase that belongs to the cytochrome P450 family [75, 86], which has monooxygenase activity (catalyzes the β-carotene hydroxylation and ketolation). However, crtS depends on crtR since the latter, due to its reductase activity, provides the necessary electrons to astaxanthin syn-thase to catalyze the substrate oxygenation [75, 86].

crtYB overexpression is essential to increase carot-enoids biosynthesis, but to overproduce astaxanthin it is necessary to overexpress astaxanthin synthase (crtS) [75]. Because of this, Ledetzky et al. [56] generated a X. dendrorhous strain containing three copies of crtYB and one of crtS, whose astaxanthin biosynthesis increased to 70% of total carotenoids produced (Fig. 2C) [57].

Nevertheless, the described increase of carotenoids synthesis is not enough to compete against the produc-tion costs of chemical synthesis. Therefore, it is necessary to explore other genes/enzymes of the metabolic pathway involved in astaxanthin biosynthesis.

Overexpression of the crtI geneThe gene crtI encodes phytoene desaturase in X. den-drorhous, the second enzyme of the carotenoid biosyn-thesis pathway (Fig.  1B). The introduction of several copies of crtI increased the accumulation of intracellu-lar lycopene [89, 91], which generated pink to dark red colonies of the yeast. Despite an increase of β-carotene and astaxanthin was expected, their concentration in fact decreased. Nevertheless, an increase of torulene and 3-hydroxy-3,4-didehydro-β, Ψ-carotene-4-ona (HDCO; Fig. 1B) was detected [89, 91]. This implies that the over expression of the phytoene desaturase in X. dendrorhous shifts the metabolic flux towards torulene biosynthesis (Fig. 1B) [80, 89].

Xie et al. [92], in order to overproduce lycopene, con-structed a S. cerevisiae strain heterologously express-ing in vitro evolved copies of crtYB, crtE and crtI genes from X. dendrorhous. They first inactivated crtYB in X. dendrorhous and, after applying directed evolution schemes to the inactivated crtYB, as well as the active crtE and crtI, they obtained mutant strains with phy-toene synthase activity able to accumulate phytoene, but decreased accumulation of β-carotene. Evolved crtYB, crtE, and crtI were introduced into S. cerevisiae to gener-ate a strain that produced more than 23 mg lycopene/g of

Table 2 Strains of X. dendrorhous and S. cerevisiae genetically engineered to overproduce carotenoids/astaxanthin

Microorganism Genetic modification scheme

Mutagenesis method Targeted gene Product Yield (mg/g) References

X. dendrorhous None Astaxanthin 0.2 to 0.4 [14, 44, 63]

Random EMS Random Astaxanthin 6.4 [4]

NTG Random Astaxanthin 4 [79, 80]

NTG Random Astaxanthin 0.5 [78]

NTG Astaxanthin 2 [82]

Site‑specific Gene duplication crtYB β‑Carotene 0.65 [89]

Gene duplication crtYB Astaxanthin 0.49 [56]

Inactivation CYP61 Astaxanthin 0.28 [88]

Gene duplication crtE, crtS β‑Carotene 0.47 [59]

Inactivation crtI/gene duplication HMG1, crtE, crtYB Phytoene 10 [58]

Inactivation crtS/gene duplication crtZ, crtYB Zeaxanthin 5.2 [61]

Gene duplication/NTG crtYB, crtS Astaxanthin 9.7 [66]

S. cerevisiae Site‑specific Heterologous integration crtYB, crtE, crtI Lycopene 23 [92]

Heterologous integration crtE, crtYB, crtI β‑Carotene 6.01 [100]

Heterologous integration/protein engineering crtE, crtI, crtYB, crtZ, bkt Astaxanthin 8.1 [96]

Page 8 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

dry biomass. Lycopene accumulation was possible since S. cerevisiae does not express astaxanthin synthase (crtS), responsible for the astaxanthin biosynthesis, HDCO and other unwanted byproducts of the astaxanthin biosyn-thesis pathway (Fig. 1B).

Similarly, Yamada et  al. [93] constructed a S. cerevi-siae strain (YTPH499/Mo3Crt79) that produced 6.01 mg β-carotene/g dry biomass in 96  h. The transformation was achieved through the simultaneous integration of three copies of crtE, one copy of crtYB and one copy of crtI. All genes were isolated from X. dendrorhous.

In accordance with the discussed results, it is necessary to find metabolic engineering alternatives that do not divert the carbon flow towards the synthesis of torulene, HDCO or unwanted byproducts but that allow the accu-mulation of astaxanthin as the main product.

Overexpression of the crtS geneAstaxanthin synthase, encoded by crtS in X. dendrorhous, together with cytochrome P450 reductase (crtR), cata-lyzes the last step of the astaxanthin pathway transform-ing β-carotene into astaxanthin. This is the limiting step in the astaxanthin biosynthesis. As previously mentioned in “Overexpression/inactivation of the crtYB gene” sec-tion, crtS has monooxygenase activity (catalyzes the β-carotene hydroxylation and ketolation) that depends on crtR activity [75, 86].

Gassel et  al. [66] carried out the integration of addi-tional copies of crtYB and crtS in X. dendrorhous genome to increase astaxanthin production (Fig. 2D). Afterwards, random chemical mutagenesis with NTG (200  µg/mL) was performed. Surviving colonies were exposed to the same treatment several rounds until the maximum asta-xanthin producer was isolated, AXJ-20/crtYB. This strain was able to produce > 9.7  mg astaxanthin/g of dry bio-mass, after isolating it in 200 µg/mL triazine-containing media. Chemical mutagenesis could alter other genes involved in astaxanthin biosynthesis [66]. For this reason, it is necessary to determine the transcriptomic and prot-eomic profiles of obtained strains.

Contreras et al. [75] generated two new X. dendrorhous strains that contained either one (Xs_1H1S) or two (Xd_2H2S) copies of crtS. While the control strain, UCD 67–385, achieved astaxanthin yields of 128 µg/g dry bio-mass, Xs_1H1S and Xd_2H2S reached a production of 154  µg/g and 179  µg/g biomass in 96  h, respectively. X. dendrorhous has been also transformed with additional copies of crtS and crtE using a Cre–lox-P system (Fig. 3). This strain, Xd-ES, was able to increase β-carotene and astaxanthin yields above 0.47  mg/g and 0.55  mg/g of dry biomass, respectively [59]. Nevertheless, the yield obtained is not yet enough to compete with astaxanthin

produced by chemical synthesis or from improved micro-bial sources obtained by random mutagenesis.

Since crtS acts together with crtR [75, 86] to generate astaxanthin, it is necessary to explore the overexpression of both genes [49, 51].

Importance of crtR in the carotenoids metabolic pathwaycrtR encodes cytochrome P450 reductase, which is involved in the last limiting step of the astaxanthin bio-synthesis pathway in X. dendrorhous [61]. As was men-tioned in “Overexpression/inactivation of the crtYB gene” section, crtR provides electrons to astaxanthin syn-thase, which catalyzes β-carotene oxygenation leading to astaxanthin synthesis [75, 86]. In order to explore the relationship of crtR with crtS and its importance in the biosynthesis of the compound, Alcaíno et  al. [86] inac-tivated crtR in two wild-type strains of X. dendrorhous, CBSTr and T13. Both strains were pale because they were unable to synthesize astaxanthin or only in mini-mal amounts, but accumulated β-carotene in concentra-tions of 114 and 50  mg/L, respectively. Parental strains were able to produce only 9 mg/L β-carotene and above 100 mg/L astaxanthin. These results revealed the impor-tance of crtR, whose knock down led to accumulation of β-carotene and concomitantly stopped the astaxanthin production [86]. Furthermore, the deletion generated in both wild-type strains affected differently as indicated by the β-carotene production obtained. These results also show that, even among wild X. dendrorhous strains, there may be a high level of polymorphism since the same genetic modification affected differently the metabolic capacity of each engineered strain [40, 86].

Some authors have stated that it is improbable to obtain X. dendrorhous overproducing mutants due to low crtR expression levels. It is necessary to integrally study the regulation of crtR and crtS, which work together in the conversion of β-carotene to astaxanthin, since they have only been studied separately [86]. Additionally, there are thirteen cytochrome P450 genes [94] whose products are monooxygenases that could be involved in both pri-mary and secondary metabolism (such as sterol biosyn-thesis, carotenoid biosynthesis and aromatic compound degradation). Alcaíno et al. [94] characterized the genes that encode other two monooxygenases belonging to cytochrome P450 family, which participate in the sterol biosynthesis: CYP51 and CYP61 (encoding for lanosterol 14-alpha demethylase and C-22 sterol desaturase, respec-tively). Since CYP51 and CYP61 are also involved in ergosterol biosynthesis, they have the ability to decrease the flux of available carbon for astaxanthin production (Fig. 1B).

Yamamoto et al. [88] reported a complete CYP61 dele-tion in a X. dendrorhous diploid strain. The result was

Page 9 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

a promising strategy to decrease ergosterol biosynthe-sis and, consequently, increase the carbon flux towards astaxanthin biosynthesis (Fig.  1B), obtaining 1.4-times higher astaxanthin production compared to the parental strain. This metabolic engineering strategy is based in the fact that ergosterol, a product of lipid biosynthesis, is an inhibitor of ERG13 and HMG1, which act in the synthesis

of mevalonate (Fig.  1B). In addition, the CYP61 disrup-tion in wild-type X. dendrorhous strains increased the HMG1 expression levels from 2 to 5 times during asta-xanthin biosynthesis [95]. It is necessary to insightfully study cytochrome P450 monooxygenases since some other genes could be related directly or indirectly to the astaxanthin metabolic pathway [57, 79, 88, 94, 95].

Fig. 3 Integration of additional copies of crtE and crtS using a Cre/lox-P system [59]. The Cre–loxP recombination system allows selectable markers recycling. Holding the transitory expression of cre (encoding Cre recombinase) controlled by an unstable vector, the selectable marker flanked by loxP sites were removed along with the Cre vector

Page 10 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

Astaxanthin biosynthesis in heterologous hostsAstaxanthin biosynthesis has been recreated in S. cerevi-siae by integrating heterologous genes encoding the nec-essary enzymatic machinery. Zhou et  al. [96] generated a S. cerevisiae strain (YastD-01) capable of producing astaxanthin in high yields (up to 8.1 mg/g dry biomass). As a first step, they built a β-carotene producing strain by heterologously expressing crtE, crtI and crtYB from X. dendrorhous. Subsequently, they expressed crtZ and bkt from H. pluvialis. crtZ encodes β-carotene hydroxy-lase, which catalyzes the conversion of canthaxanthin to astaxanthin. bkt encodes β-carotene ketolase, which cata-lyzes the conversion of β-carotene to canthaxanthin [96]. The insertion of both H. pluvialis genes led to a yield of 8.1 mg astaxanthin/g of biomass, but it turned out to be the (3S, 3′S) stereoisomer. The graphic representation of this genetic engineering scheme is shown in Fig.  4. Similarly, Jiang et al. [97] generated a recombinant S. cer-evisiae strain by introducing the carRA (orthologous to crtYB from X. dendrorhous) and crtI genes from Blakes-lea trispora, crtZ from Agrobacterium aurantiacum, crtW from Brevundimonas vesicularis, which encodes a β-carotene ketolase (non-encoded in X. dendrorhous

genome) and crtE from Taxus media. The transforming S. cerevisiae strain was only capable of producing 20 mg astaxanthin/L, but, additionally, it was subjected to sev-eral rounds of physical mutagenesis and adaptation to hydrogen peroxide. After 15 rounds, the best selected strain was able to produce > 65  mg astaxanthin/L in a batch system, while in a fed batch system the maximum concentration obtained was 404.78 mg/L.

In terms of (3R, 3′R)-astaxanthin overproduction, it has not yet been achieved in satisfactory levels. Table  2 shows a summary of the most relevant works on meta-bolic engineering for astaxanthin biosynthesis, both in X. dendrorhous and S. cerevisiae. The highest astaxanthin yield to date is around 10  mg/g biomass. Since X. den-drorhous is the microorganism that possesses the native enzymatic machinery for the astaxanthin biosynthesis, it could be the option with the greatest potential for meta-bolic engineering. However, it is still necessary to study in greater detail the carbon flux and the metabolic limita-tions that exist in this biological system, the goal we are pursuing in this review.

Metabolic strategies to increase the carbon flux to astaxanthin biosynthesis in X. dendrorhousNowadays, metabolic engineering of X. dendrorhous has not led to competitive astaxanthin (3R, 3′R) overproduc-ing mutants. In addition to the use of mevalonate, citrate, glutamate, succinate, glycerol and other molecules as inductors of enzymes that participate in the astaxanthin biosynthesis [30, 74, 98, 99], it is necessary to establish the influence of sugars metabolism and the tricarboxylic acids cycle in the carbon flow towards the carotenoids biosynthesis [48, 49]. This information can be used to identify enzymes that participate in those metabolic pathways that increase the flux towards astaxanthin accumulation.

In species such as Candida utilis, S. cerevisiae and other yeasts, a strong limitation in the expression levels of HMG-CoA reductase and GGPP synthase has been found and, subsequently, are limited the levels of meva-lonate and isoprenoids derivatives available for carot-enoid and astaxanthin biosynthesis [30, 31, 35, 49, 100]. Analogously, this could be one of the reasons why the carbon flux in X. dendrorhous could be reduced for iso-prenoids, lipids and carotenoid biosynthesis (Fig. 1B).

In X. dendrorhous, phytoene-β-carotene synthase is a key factor to displace the metabolic flux to astaxanthin biosynthesis from its substrate, GGPP [30, 31, 33]. If there is not enough substrate for astaxanthin biosynthe-sis, the carbon flux shifts to isoprenoids and sterols bio-synthesis (Fig. 1B) [30, 31, 42]. It has been demonstrated that the inactivation of the gene that encodes for squalene synthase in C. utilis and, in parallel, overexpression of

Fig. 4 Metabolic engineering in S. cerevisiae for astaxanthin (3S, 3′S) overproduction [96]. A β‑carotene overproducer S. cerevisiae strain was obtained by heterologously expressing crtE, crtI and crtYB genes from X. dendrorhous. Subsequently, crtZ and bkt from the microalgae H. pluvialis were expressed. crtZ encodes for β‑carotene hydroxylase which catalyzes the conversion of canthaxanthin to astaxanthin. bkt encodes for β‑carotene ketolase which catalyzes the conversion of β‑carotene to canthaxanthin

Page 11 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

HMG-CoA reductase, increases the carotenoid biosyn-thesis up to 2 times [30, 31, 35]. Similarly in S. cerevi-siae, decreasing the regulation of ERG20 by inactivation of squalene synthase and increasing GGPP synthesis by overexpression of crtE, causes an increase in carot-enoids synthesis [31]. Moreover, as mentioned earlier in “Importance of crtR in the carotenoids metabolic path-way” section, Yamamoto et al. [88] demonstrated that the complete deletion of CYP61 (C-22 sterol desaturase) in X. dendrorhous increases HMG1 expression levels from 2 to 5 times [95] and, subsequently, increases the carbon flux towards astaxanthin biosynthesis (production above 1.4 times). This is due to the decrease of ergosterol bio-synthesis, which is an inhibitor of ERG13 and HMG1 activities (Fig.  1B). In addition, CYP61 deletion in X. dendrorhous leads to an increased SRE1 expression level (around 3.5 times compared with wild-type strains) [101] with which the metabolic pathways derived from terpe-noid biosynthesis are activated, and thus astaxanthin and ergosterol precursors production.

Nevertheless, high yields of carotenoids may affect the cell since their intracellular accumulation could be toxic [49, 52]. However, increasing the ability of the host to accumulate big amounts of lipids is possible [102] using directed evolution techniques [32] or overexpressing genes that encode enzymes involved in the cell mem-brane flexibility. In S. cerevisiae, it has been observed that overexpression of stearoyl-CoA 9-desaturase increases the capacity to accumulate lycopene up to 30% [103], while overexpressing ino2, which encodes a transcrip-tional factor involved in the response to physiological stress, increases up to 10% the ability to accumulate lipids and derivatives (including lycopene) [104].

To determine which are the metabolic mechanisms that influence the carotenoids and astaxanthin biosyn-thesis in X. dendrorhous, proteomic and transcriptomic analysis have been performed on overproducing strains [76, 81]. Barbachano-Torres et  al. [76], using random mutagenesis with NTG on a wild-type strain, obtained an XR4 mutant able to synthesize 2.5–8.8 times more asta-xanthin than the parental strain. The proteomic profile of the mutant was 90% identical to the wild-type strain; however, they found a decrease in NADP(H)-dependent glutamate dehydrogenase 1 (GDH1) and UDP-glucuronic acid decarboxylase; an increase of NADP(H)-dependent glutamate dehydrogenase 3 (GDH3), glyceraldehyde 3-phosphate dehydrogenase (GPDH), dihydrolipoyl dehydrogenase and a probable pyridoxine synthase. GDH1 and GDH3 are involved in ammonium assimila-tion and glutamate biosynthesis, respectively [43, 105, 106]. Glutamate has recently been shown to aid in carbon and nitrogen assimilation, which increases astaxanthin biosynthesis [106, 107]. On the other hand, dihydrolipoyl

dehydrogenase participates in the metabolism of tricar-boxylic acids, which is why it promotes the generation of reactive oxygen species and the subsequent biosynthe-sis of carotenoids in response to oxidative damage [76, 108]. GPDH is related to the production of high-energy molecules during glycolysis, and it is likely that other oxidation reactions are triggered including those that lead to cellular apoptosis. In response to this oxidative damage, genes involved in astaxanthin biosynthesis are transcribed [44, 51]. UDP-glucuronic acid decarboxy-lase is related to the metabolism and assimilation of sugars, which is why it maintains the cellular integrity [76, 109]. It is probable that the overexpression of these enzymes could play an important role in the generation of overproducing strains. A transcriptomic analysis of XR4 mutant strain (capable of overproducing 2.5–8.8 times more astaxanthin than the wild-type strain) [76, 81] revealed that it overexpresses crtE and crtS, and crtI at levels 3 and 2 times above the control strain, respec-tively. However, expression levels of genes such as idi were variable, while crtYB and crtR expression was even lower than those of the control strain. According to the expression levels of the genes involved in the carotenoids and astaxanthin biosynthesis, crtS action is essential to astaxanthin biosynthesis while crtR could be expressed at non-limiting levels in this overproducer X. dendrorhous strain. Nevertheless, the crtR expression levels may be variable and limiting between different mutant strains due to their polymorphism [40, 86] and different capaci-ties to displace carbon flux towards the different meta-bolic pathways [84].

Pan et  al. [106] demonstrated that using a 76:1 C/N ratio is critical to increase the concentration of astax-anthin in X. dendrorhous UV3-721 mutant strain. After a proteomic analysis, they found 1.5- to 2.5-fold more Grg2, aspartic protease precursor, transaldolase, 3-iso-propylmalate dehydrogenase and a disulfide isomerase precursor compared to that observed at low 19:1 C/N ratio. They also observed sixfold overexpression of phos-pho-methyl-pyridine kinase, 3.3 times more astaxan-thin synthase, and overexpression of IPP isomerase and pyruvate decarboxylase. On the other hand, in this study, unlike the results reported by Barbachano et  al. [76], a tenfold decrease in GPDH expression was observed. This may be due to the high C/N ratio used as a metabolic stressor to induce the generation of carotenoids instead of high energy molecules in the tricarboxylic acid cycle. In addition, overexpression of other key enzymes such as pyruvate decarboxylase, IPP isomerase and astaxan-thin synthase, strongly correlated with the biosynthesis of carotenoid precursors and astaxanthin [51, 59].

The strategy of feeding 0.368  g/L glutamate in media culture, preserving a high 76:1 C/N ratio, has been used

Page 12 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

[74]. Glutamate participates in the tricarboxylic acids cycle in the form of 2-ketoglutarate due to the activity of GDH. Because of that, it is an activator of carbon and nitrogen assimilation, which provides energy to the cell and leads to a 40% increase in astaxanthin biosynthesis [74, 76]. Under these conditions, a proteomic analysis showed a 2–2.5-fold overexpression of phytoene desatu-rase, GDH, glycerate and formate dehydrogenase, IPP dehydrogenase and cytochrome P450 reductase. There was also a 1.3–1.5-fold overexpression of dihydrokaemp-ferol 4-reductase, 3-isopropyl malate dehydrogenase and astaxanthin synthase. These data were corroborated by assessing the corresponding expression levels. All these enzymes play a key role in the primary and secondary metabolism, and evidently influence the astaxanthin bio-synthesis. This is why it is important to consider them in the metabolic engineering processes to generate astaxan-thin overproducing strains [74].

Proteomic analysis has revealed that the oxida-tive stress generated by the tricarboxylic acid cycle is a key factor in increasing the expression levels of genes involved in astaxanthin biosynthesis, as well as of pre-cursor compounds [51, 74, 76, 106]. The use of succinate (20  g/L) as a carbon source has been used by Wozniak et  al. [98] to induce astaxanthin biosynthesis, obtaining 2.5 times more astaxanthin compared to that obtained when glucose (20 g/L) is used as carbon source. Through a transcriptional analysis, it was shown that succinate, after the exponential growth phase, increased the expres-sion levels of crtYB (4.2- to 4.5-fold), crtI (1.7- to 1.8-fold) and crtS (up to 1.3-fold) compared to the use of glucose as carbon source [98].

Moreover, decreasing the expression levels of Mig1 in X. dendrorhous leads to twofold increase of crtYB and crtS expression levels [62]. However, it was observed that the expression of crtI in the mutant strain was lower than that of the wild-type one (− 2 times). Since crtS and crtR work simultaneously in the astaxanthin bio-synthesis (Fig. 1B), it has been reported that crtS overex-pression and maintaining non-limiting crtR expression levels [76, 86], resulted in 5.5-fold increase of astaxanthin biosynthesis.

Mig1 is a catabolic repressor that binds to specific DNA sequences called “Mig1 boxes” in the promoter regions of various glucose-repressed genes in yeast [62, 98]. Mig1 union sites in crtYB and crtI have been identified, while two sites were detected on the regulatory region of crtS gene [98]. Mig1 is regulated by phosphorylation. In the absence of glucose, Mig1 is phosphorylated by Snf1 kinase, and allocated to the cytoplasm. However, in the presence of glucose, an intracellular signaling cascade activates the dephosphorylation of Mig1, which sub-sequently migrates towards the nucleus where it binds

to the transcriptional corepressor complex [110, 111]. Therefore, decreasing the Mig1 expression levels repre-sents an alternative to increase the carbon flow towards the synthesis of astaxanthin in X. dendrorhous.

Taking advantage of cell division acceleration induced by plant cytokinin 6-benzylaminopurine (6-BAP) [112], Pan et  al. [51] managed to increase biomass (up to 3.33 g/L) and astaxanthin production (4.67 mg/g dry bio-mass). The metabolome of 6-BAP-treated yeast revealed glycolysis induction (although the rate of carbohydrate consumption decreased), and suppression of the tricar-boxylic acids cycle that prevented diverting the carbon flow to unwanted metabolic routes. 6-BAP also triggered oxidative stress leading to increased trehalose biosyn-thesis (from glucose) and accumulation of astaxanthin in response to the high concentration of reactive oxygen species [51, 113, 114].

Transcriptional analysis of 6-BAP treated X. den-drorhous showed overexpression of astaxanthin biosyn-thesis genes [51]. HMG1 and idi showed a significant 3.5-fold overexpression, while crtE had a twofold increase. crtS and crtYB had a 1.7 and twofold increase, respectively [51]. crtI, in the middle of the fermentation process, had a 1.35-fold expression increase, but at the end it decreased to its basal level [51]. It is important to note that it is necessary to maintain crtI expression lev-els close to basal since increasing its activity can divert the carbon flux towards the generation of torulene and HDCO [89, 91]. Regarding maximization of carbon flux towards mevalonate and carotenoids biosynthesis, mod-ulation of all these genes by 6-BAP significantly increased the substrate availability for subsequent accumulation of astaxanthin [53].

The synthesis of HMGR and NADPH is also important in the mevalonate metabolic pathway for the synthesis of carotenoid precursors [49, 115–117]. Within the possi-ble scenarios of metabolic engineering in the tricarbox-ylic acid cycle, overproduction of NADPH [49, 50, 118], through the overexpression of citrate synthase, succinate dehydrogenase and glutamate dehydrogenase, in addition to the overproduction of malate dehydrogenase, could increase the accumulation of carotenoid precursors and push the carbon flux towards astaxanthin biosynthesis in X. dendrorhous.

The X. dendrorhous Sre1 functions as a sterol regu-latory element-binding protein (SREBP) [101, 119]. The transcription factor domain of Sre1 (the N-ter-minal domain) has a DNA binding motif [120] and the C-terminal domain (regulatory domain) interacts with a protein called SCAP (SREBP cleavage activat-ing protein, named Scp1 in S. cerevisiae) that inhibits sterols biosynthesis when the cellular levels are suffi-cient. When sterol levels drop, the Sre1–Scp1 complex

Page 13 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

is transported to the Golgi apparatus, leading to Sre1 release, and translocation to the nucleus. Then, tran-scription of genes involved in sterol biosynthesis is activated [121], and concomitantly carotenoids biosyn-thesis decreases [86]. However, SRE1 deletion inhib-its both sterol and carotenoid biosynthesis [101, 119]. Gutierrez et  al. [119] showed the N-terminal domain overexpression of SRE1 increases 1.5 and twofold the expression of ERG13 and HMG1. This could be used as an alternative to the isoprenoids overproduction to increase the astaxanthin biosynthesis.

In recombinant carotenoids-producing S. cerevisiae, mevalonate kinase (ERG12) and farnesyl pyrophosphate synthase (ERG20) have been recently identified as limit-ing factors for the isoprenoid accumulation in carotenoid biosynthesis [122]. Using CRISPR/Cas9 [32, 122], a S. cer-evisiae strain able to accumulate up to 11 times more iso-prenoids was obtained. ERG12 and ERG20, in a similar way to S. cerevisiae, could generate a bottleneck for the astaxanthin biosynthesis in X. dendrorhous.

The genomic, transcriptomic and proteomic data described above, indicate it is necessary to increase the carbon flux towards astaxanthin biosynthesis. Knowing that the mevalonate pathway and its isoprenoid prod-ucts feed carotenoid biosynthesis [17, 20, 49, 51], it is necessary to simultaneously overexpress HMG1, ERG12, ERG20 and idi genes that encode enzymes that limit the accumulation of FPP, the main substrate for carotenoid biosynthesis. Despite the latter has been performed in S. cerevisiae and Y. lipolytica, this metabolic engineering program has not been carried out in X. dendrorhous. If increasing FPP accumulation is achieved, a second step focused on increasing the expression of genes directly involved in the astaxanthin biosynthesis pathway is nec-essary [33, 51, 55–59, 61]. The low activity of crtYB, crtR and crtS are the main bottlenecks in the synthesis of asta-xanthin and their overexpression would also prevent the accumulation of isoprenoid products, thus avoiding the effect of inhibition by products such as sterols.

Among other important considerations, decrease of catabolic repression in X. dendrorhous activates the entire metabolic machinery for carotenoid/astaxanthin biosynthesis, a process that results from the inactivation of MIG1 and, subsequently, increase of crtYB and crtS expression levels [62, 110, 111].

Overexpressing stearoyl-CoA 9-desaturase (OLE1) has been shown to increase the accumulation of substrates for carotenoid biosynthesis [103] or, in contrast, deletion of CYP61 increases the substrates availability for astaxan-thin biosynthesis and inhibition by sterols is evaded [88]. On the other hand, overexpressing enzymes involved in the tricarboxylic acid cycle such as NADPH oxidase, cit-rate synthase, succinate dehydrogenase and glutamate

dehydrogenase may increase the carbon flux [49–51, 118] to astaxanthin accumulation.

The presence of oxygen, including reactive oxygen spe-cies such as H2O2, stimulates the production of carot-enoids and astaxanthin as a protection mechanism in response to oxidative cell damage [69, 123–126]. Due to the above, it is proposed to turn off the enzymatic machinery in charge of evading reactive oxygen species so that carotenoids/astaxanthin biosynthesis be the main antioxidative response. Therefore, turning off CTT1, which encodes for catalase in X. dendrorhous, as well as knocking out the genes that encode for superoxide dis-mutase (SOD1) and glutathione peroxidase, may further induce astaxanthin biosynthesis. In X. dendrorhous, four copies of CTT1 have been found [126]. In S. cerevisiae, it has been found a great diversity of glutathione peroxi-dases encoded by the genes URE2, GPX1, GPX2, GTT1 and/or hyr1p [126] able to prevent intracellular oxidative damage.

Another potentially powerful strategy to favor second-ary metabolism in X. dendrorhous and thereby promote the biosynthesis of carotenoids and astaxanthin is to keep the biological system under constant oxidative stress throughout the fermentation process. Taking advantage of the knowledge gained about the addition of H2O2 as an inducer of the production of astaxanthin [123], it is possi-ble to obtain a genetically modified strain able to produce H2O2 autonomously. Within the genome of basidiomy-cota fungi, the superfamily of glucose-methanol-choline (GMC) oxidoreductases includes enzymes that produce extracellular H2O2 involved in the degradation of struc-tural plant polysaccharides such as lignin, cellulose and hemicellulose [127, 128]. Within this superfamily can be found glucose oxidase (GOX), glucose dehydrogenase (GDH), ethanol/methanol oxidase (MOX), aryl-alcohol oxidase (AAO), pyranose 2-oxidase (P2O), pyranose dehydrogenase (PDH) and glyoxal oxidase (GLX). Het-erologously expressing some of these genes in X. den-drorhous and, in addition, knocking out the reductases (catalase, superoxide dismutase and glutathione peroxi-dase) necessary to avoid oxidative damage, could favor the overproduction of astaxanthin.

ConclusionsAs discussed above, there are a number of different strat-egies to obtain genetically engineered X. dendrorhous strains for carotenoids production in high yields [32, 49, 56, 59, 61, 66]. Nevertheless, β-carotene and (3R, 3′R)-astaxanthin yields are still low. It is obviously necessary to generate more knowledge that addresses the limita-tions of the astaxanthin biosynthesis pathway [51, 62, 74].

Random mutagenesis using physical and chemical agents [8, 29, 66, 71, 78] have led to X. dendrorhous strains

Page 14 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

increase astaxanthin production [66]. Nevertheless, the targeted genes are usually unknown, which represents a disadvantage for the development of new metabolic engi-neering schemes to overproduce (3R, 3′R)-astaxanthin.

Genes involved in the carotenoids pathway have been induced to increase astaxanthin biosynthesis, but the results have not been enough. It is necessary to inactivate other genes such CYP61 and/or MIG1 (the latter involved in catabolic repression) and, in addition, induce protein domains such as SRE1 N-terminal, indirectly related to the metabolism of carotenoids/astaxanthin [62, 95, 110, 111, 120, 121]. However, it is necessary to evaluate the multi-ple transformations directly on X. dendrorhous to increase the (3R, 3′R)-astaxanthin production. The improvement of astaxanthin overproduction is a request.

New metabolic engineering strategies on X. dendrorhous (such as those proposed in this review and others) could be implemented and studied to obtain more and new spe-cific information of other regulatory mechanisms involved directly or indirectly to astaxanthin biosynthesis and, with this, achieve a genetic modification strategy capable of pro-ducing the compound in high yields.

AcknowledgementsWe thank the valuable comments from an anonymous reviewer that helped improve this manuscript.

Authors’ contributionsAT‑H collected data and drafted the manuscript. AT‑H, JV, MRK and MA‑P analyzed and discussed data, and prepared the final version of the manuscript. All authors read and approved the final manuscript.

FundingThis work was supported by COECYTJAL‑Mexico (9141‑2020). Alejandro Torres Haro was the recipient of a Ph.D. scholarship from CONACYT‑Mexico (783291).

Availability of data and materialsNot applicable.

Declarations

Ethics approval and consent to participateThe authors declare having consented to participate in the work and carried it out under professional ethics standards.

Consent for publicationAll authors agreed to submit the work for publication.

Competing interestsThe authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Received: 18 March 2021 Accepted: 23 August 2021

References 1. Mata‑Gómez LC, Montañez JC, Méndez‑Zavala A, Aguilar CN. Biotech‑

nological production of carotenoids by yeasts: an overview. Microb Cell Fact. 2014;13(1):1–11.

2. Gio‑Bin L, Sang‑Yun L, Eun‑Kyu L, Seung‑Joo H, Woo‑Sik K. Separa‑tion of astaxanthin from red yeast Phaffia rhodozyma by supercritical carbon dioxide extraction. Biochem Eng J. 2002;11(2–3):181–7.

3. Bhuvaneswari S, Arunkumar E, Viswanathan P, Anuradha CV. Astax‑anthin restricts weight gain, promotes insulin sensitivity and curtails fatty liver disease in mice fed a obesity‑promoting diet. Process Biochem. 2010;45(8):1406–14.

4. Gui‑Li J, Ling‑Yan Z, Yu‑Tao W, Ming‑Jun Z. Astaxanthin from Jerusalem artichoke: production by fed‑batch fermentation using Phaffia rhodozyma and application in cosmetics. Process Biochem. 2017;63:16–25.

5. Wang HD, Chen C, Huynh P, Chang J. Exploring the potential of using algae in cosmetics. Bioresour Technol. 2015;184:355–62.

6. Avalos J, Carmen LM. Biological roles of fungal carotenoids. Curr Genet. 2015;61(3):309–24.

7. Bhosale P. Environmental and cultural stimulants in the production of carotenoids from microorganisms. Appl Microbiol Biotechnol. 2004;63(4):351–61.

8. Kim JH, Kang SW, Kim SW, Chang HI. High‑level production of astaxanthin by Xanthophyllomyces dendrorhous mutant JH1 using statistical experimental designs. Biosci Biotechnol Biochem. 2005;69(9):1743–8.

9. Mortensen A, Skibsted LH, Truscott TG. The interaction of dietary carot‑enoids with radical species. Arch Biochem Biophys. 2001;385(1):13–9.

10. Naguib YMA. Antioxidant activities of astaxanthin and related carot‑enoids. J Agric Food Chem. 2000;48(4):1150–4.

11. Saini RK, Keum YS. Progress in microbial carotenoids production. Indian J Microbiol. 2017;57(1):129–30.

12. Zhang C, Chen X, Too HP. Microbial astaxanthin biosynthesis: recent achievements, challenges, and commercialization outlook. Appl Micro‑biol Biotechnol. 2020;104(13):5725–37.

13. Domínguez‑Bocanegra AR, Ponce‑Noyola T, Torres‑Muñoz JA. Astaxanthin production by Phaffia rhodozyma and Haematococcus pluvialis: a comparative study. Appl Microbiol Biotechnol Biotechnol. 2007;75:783–91.

14. Higuera‑Ciapara I, Félix‑Valenzuela L, Goycoolea FM. Astaxanthin: a review of its chemistry and applications. Crit Rev Food Sci Nutr. 2006;46(2):185–96.

15. Olaizola M, Huntley ME. Recent advances in commercial production of astaxanthin from microalgae. Biomater Bioprocess. 2003;9:143–64.

16. Bernhart K. Synthetic astaxanthin. The route of a carotenoid from research to commercialisation. In: Krinsky NI, MathewsRoss MM, Taylor RF, editors. Carotenoids: chemistry and biology. Boston: Springer; 1989. p. 337–63.

17. Foss P, Storebakken T, Austreng E, Liaaenjensen S. Carotenoids in diets for salmonids. Aquaculture. 1987;65(3–4):293–305.

18. Storebakken T, Foss P, Schiedt K, Austreng E, Liaaen‑Jensen S, Manz U. Carotenoids in diets for salmonids. IV. Pigmentation of Atlantic salmon with astaxanthin, astaxanthin dipalmitate and canthaxanthin. Aquacul‑ture. 1987;65(3–4):279–92.

19. Østerlie M, Bjerkeng B, Liaaen‑Jensen S. Plasma appearance and distribution of astaxanthin E/Z and R/S isomers in plasma lipoproteins of men after single dose administration of astaxanthin. J Nutr Biochem. 2000;11(10):482–90.

20. Andrewes AG, Phaff HJ, Starr MP. Carotenoids of Phaffia rhodozyma, a red‑pigmented fermenting yeast. Phytochemistry. 1976;15(6):1003–7.

21. Boussiba S. Procedure for large‑scale production of astaxanthin from Hematococcus. United States; 6022701; 2000. https:// paten ts. google. com/ patent/ US602 2701A/ en.

22. Sun W, Xing L, Lin H, Leng K, Zhai Y, Liuet X. Assessment and compari‑son of in vitro immunoregulatory activity of three astaxanthin stereoi‑somers. J Ocean Univ China. 2016;15:283–7.

23. Nakano T, Kanmuri T, Sato M, Takeuchi M. Effect of astaxanthin rich red yeast (Phaffia rhodozyma) on oxidative stress in rainbow trout. Biochim Biophys Acta. 1999;1426:119–25.

24. Schroeder WA, Johnson EA. Antioxidant role of carotenoids in Phaffia rhodozyma. Microbiology. 1993;139:907–12.

25. Takimoto T, Takahashi K, Akiba Y. Effect of dietary supplementation of astaxanthin by Phaffia rhodozyma on lipid peroxidation, drug metabo‑lism and some immunological variables in male broiler chicks fed on diets with or without oxidised fat. Br Poult Sci. 2007;48:90–7.

Page 15 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

26. Amado IR, Vázquez JA. Mussel processing wastewater: a low‑cost substrate for the production of astaxanthin by Xanthophyllomyces dendrorhous. Microb Cell Fact. 2015;14(1):1–11.

27. Sánchez‑Ortiz AF. Monitoreo del estado fisiológico de la levadura Xan-thophyllomyces dendrorhous durante la producción de astaxantina por medio de técnicas espectroscópicas y análisis de imágenes. CIATEJ, AC; 2015. https:// ciatej. repos itori oinst ituci onal. mx/ jspui/ handle/ 1023/ 54.

28. Kim DK, Hong SJ, Bae JH, Yim N, Jin E, Lee CG. Transcriptomic analysis of Haematococcus lacustris during astaxanthin accumulation under high irradiance and nutrient starvation. Biotechnol Bioprocess Eng. 2011;16(4):698–705.

29. Ramírez J, Obledo N, Arellano M, Herrera E. Astaxanthin production by Phaffia rhodozyma in a fedbatch culture using a low cost medium feed‑ing. Rev Electrónica y Tecnológica e‑Gnosis. 2006;4. www.e‑ gnosis. udg. mx/ vol4/ art5.

30. Schmidt I, Schewe H, Gassel S, Jin C, Buckingham J, Hümbelin M, Sand‑mann G, Schrader J. Biotechnological production of astaxanthin with Phaffia rhodozyma/Xanthophyllomyces dendrorhous. Appl Microbiol Biotechnol. 2011;89(3):555–71.

31. Wan X, Zhou XR, Moncalian G, Su L, Chen WC, Zhu HZ, Chen D, Gong YM, Huang HH, Deng QC. Reprogramming microorganisms for the biosynthesis of astaxanthin via metabolic engineering. Prog Lipid Res. 2021;81:101083.

32. Guirimand G, Kulagina N, Papon N, Hasunuma T, Courdavault V. Innova‑tive tools and strategies for optimizing yeast cell factories. Trends Biotechnol. 2020;39:1–17.

33. Visser H, Van Ooyen AJJ, Verdoes JC. Metabolic engineering of the astaxanthin‑biosynthetic pathway of Xanthophyllomyces dendrorhous. FEMS Yeast Res. 2003;4(3):221–31.

34. Bhataya A, Schmidt‑Dannert C, Lee PC. Metabolic engineering of Pichia pastoris X‑33 for lycopene production. Process Biochem. 2009;44(10):1095–102.

35. Misawa N, Shimada H. Metabolic engineering for the production of carotenoids in non‑carotenogenic bacteria and yeasts. J Biotechnol. 1998;59(3):169–81.

36. Miki W. Biological functions and activities of animal carotenoids. Pure Appl Chem. 1991;63(1):141–6.

37. Golubev WI. Perfect state of Rhodomyces dendrorhous (Phaffia rho-dozyma). Yeast. 1995;11(2):101–10.

38. Bellora N, Moliné M, David‑Palma M, Coelho MA, Hittinger CT, Sampaio JP, Gonçalves P, Libkind D. Comparative genomics provides new insights into the diversity, physiology, and sexuality of the only industrially exploited tremellomycete: Phaffia rhodozyma. BMC Genom. 2016;17(1):1–16.

39. Kucsera J, Pfeiffer I, Takeo K. Biology of the red yeast Xanthophyllomyces dendrorhous (Phaffia rhodozyma). Mycoscience. 2000;41(3):195–9.

40. Kucsera J, Pfeiffer I, Ferenczy L. Homothallic life cycle in the diploid red yeast Xanthophyllomyces dendrorhous (Phaffia rhodozyma). Antonie van Leeuwenhoek, Int J Gen Mol Microbiol. 1998;73(2):163–8.

41. Martínez C, Hermosilla G, León R, Pincheira G, Cifuentes V. Genetic transformation of astaxanthin mutants of Phaffia rhodozyma. Antonie van Leeuwenhoek, Int J Gen Mol Microbiol. 1998;73(2):147–53.

42. Sharma R, Gassel S, Steiger S, Xia X, Bauer R, Sandmann G, Thines M. The genome of the basal agaricomycete Xanthophyllomyces dendrorhous provides insights into the organization of its acetyl‑CoA derived pathways and the evolution of Agaricomycotina. BMC Genom. 2015;16(1):1–13.

43. Rodríguez‑Sáiz M, De La Fuente JL, Barredo JL. Xanthophyllomyces dendrorhous for the industrial production of astaxanthin. Appl Microbiol Biotechnol. 2010;88(3):645–58.

44. Chávez‑Cabrera C. Una Vista Integral de la Síntesis de Astaxantina en Phaffia rhodozyma. BibliotecaCinvestavMx. 2010;14(3):24–38. http:// bibli oteca. cinve stav. mx/ indic adores/ texto_ compl eto/ cinve stav/ 2010/ 200676_ 1. pdf.

45. Garcia A. Study of astaxanthin production by Xanthophyllomyces den-drorhous. España: Universidad de León; 2012.

46. Johnson EA, Schroeder WA. Microbial carotenoids yield of cells on substrate. Adv Biochem Eng. 1995;53:119–78.

47. Lodato P, Alcaíno J, Barahona S, Niklitschek M, Carmona M, Wozniak A, Baeza M, Jimenez A, Cifuentes V. Expression of the carotenoid

biosynthesis genes in Xanthophyllomyces dendrorhous. Biol Res. 2007;40(1):73–84.

48. Wang J, Niyompanich S, Tai YS, Wang J, Bai W, Mahida P, Gao T, Zhang K. Engineering of a highly efficient Escherichia coli strain for meva‑lonate fermentation through chromosomal integration. Appl Environ Microbiol. 2016;82(24):7176–84.

49. Li C, Swofford CA, Sinskey AJ. Modular engineering for microbial production of carotenoids. Metab Eng Commun. 2020;10:e00118.

50. Zhao J, Li Q, Sun T, Zhu X, Xu H, Tang J, Zhan X, Ma Y. Engineer‑ing central metabolic modules of Escherichia coli for improving Β‑carotene production. Metab Eng. 2013;17(1):42–50.

51. Pan X, Wang B, Duan R, Jia J, Li J, Xiong W, Ling X, Chen C, Huang X, Zhang G, Lu Y. Enhancing astaxanthin accumulation in Xanthophyl-lomyces dendrorhous by a phytohormone: metabolomic and gene expression profiles. Microb Biotechnol. 2020;13(5):1446–60.

52. Sung WS, In‑Seon L, Dong GL. Damage to the cytoplasmatic membrane and cell death caused by lycopene in Candida albicans. J Microbiol Biotechnol. 2007;17(11):1797–804.

53. Pitera DJ, Paddon CJ, Newman JD, Keasling JD. Balancing a heterolo‑gous mevalonate pathway for improved isoprenoid production in Escherichia coli. Metab Eng. 2007;9(2):193–207.

54. Hinson D, Chambliss KL, Toth MJ, Tanaka RD, Gibson KM. Post‑translational regulation of mevalonato kinase by intermediates of the cholesterol and nonsterol isopreno biosynthetic pathways. J Lipid Res. 1997;38:2216–23.

55. Verdoes JC, Sandmann G, Visser H, Diaz M, Van MM. Metabolic engi‑neering of the carotenoid biosynthetic pathway in the yeast. Society. 2003;69(7):3728–38.

56. Ledetzky N, Osawa A, Iki K, Pollmann H, Gassel S, Breitenbach J, Shindo K, Sandmann G. Multiple transformation with the crtYB gene of the limiting enzyme increased carotenoid synthesis and generated novel derivatives in Xanthophyllomyces dendrorhous. Arch Biochem Biophys. 2014;545:141–7.

57. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual, 4th edition. Cold Spring Harbor Laboratory Press; 1989. p. 1–16. http:// coolb ook. us/ molec ular‑ cloni ng‑ a‑‑ 19361 13422. html.

58. Pollmann H, Breitenbach J, Sandmann G. Development of Xantho-phyllomyces dendrorhous as a production system for the colorless carotene phytoene. J Biotechnol. 2017;247:34–41.

59. Zhang N, Li J, Li F, Wang S. Selectable marker recycling in the nonconventional yeast Xanthophyllomyces dendrorhous by transient expression of Cre on a genetically unstable vector. Appl Microbiol Biotechnol. 2019;103(2):963–71.

60. Ukibe K, Hashida K, Yoshida N, Takagi H. Metabolic engineering of Saccharomyces cerevisiae for astaxanthin production and oxidative stress tolerance. Appl Environ Microbiol. 2009;75(22):7205–11.

61. Breitenbach J, Pollmann H, Sandmann G. Genetic modification of the carotenoid pathway in the red yeast Xanthophyllomyces den-drorhous: engineering of a high‑yield zeaxanthin strain. J Biotechnol. 2019;289:112–7.

62. Alcaíno J, Bravo N, Córdova P, Marcoleta AE, Contreras G, Barahona S, Sepúlveda D, Fernandéz‑Lobato M, Baeza M, Cifuentes V. The involve‑ment of Mig1 from Xanthophyllomyces dendrorhous in catabolic repression: an active mechanism contributing to the regulation of carotenoid production. PLoS ONE. 2016;11(9):1–24.

63. Johnson EA. Phaffia rhodozyma: colorful odyssey. Int Microbiol. 2003;6(3):169–74.

64. Tinoi J, Rakariyatham N, Deming RL. Utilization of mustard waste iso‑lates for improved production of astaxanthin by Xanthophyllomyces dendrorhous. J Ind Microbiol Biotechnol. 2006;33(4):309–14.

65. Johnson EA, Lewis MJ. Astaxanthin formation by the yeast Phaffia rhodozyma. J Gen Microbiol. 1979;115(1):173–83.

66. Gassel S, Schewe H, Schmidt I, Schrader J, Sandmann G. Multiple improvement of astaxanthin biosynthesis in Xanthophyllomyces dendrorhous by a combination of conventional mutagenesis and metabolic pathway engineering. Biotechnol Lett. 2013;35(4):565–9.

67. Niamnuy C, Devahastin S, Soponronnarit S, Vijaya Raghavan GS. Kinetics of astaxanthin degradation and color changes of dried shrimp during storage. J Food Eng. 2008;87(4):591–600.

Page 16 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

68. Meyer PS, Du Preez JC. Effect of culture conditions on astaxanthin production by a mutant of Phaffia rhodozyma in batch and chemostat culture. Appl Microbiol Biotechnol. 1994;40(6):780–5.

69. Yamane Y, Higashida K, Nakashimada Y, Kakizono T, Nishio N. Influ‑ence of oxygen and glucose on primary metabolism and astax‑anthin production by Phaffia rhodozyma in batch and fed‑batch cultures: kinetic and stoichiometric analysis. Appl Environ Microbiol. 1997;63(11):4471–8.

70. Ramírez J, Gutierrez H, Gschaedler A. Optimization of astaxanthin pro‑duction by Phaffia rhodozyma through factorial design and response surface methodology. J Biotechnol. 2001;88(3):259–68.

71. Liu YS, Wu JY. Optimization of cell growth and carotenoid production of Xanthophyllomyces dendrorhous through statistical experiment design. Biochem Eng J. 2007;36(2):182–9.

72. An GH, Johnson EA. Influence of light on growth and pigmenta‑tion of the yeast Phaffia rhodozyma. Antonie Van Leeuwenhoek. 1990;57(4):191–203.

73. Wu W, Lu M, Yu L. Expression of carotenogenic genes and astaxanthin production in Xanthophyllomyces dendrorhous as a function of oxygen tension. Zeitschrift fur Naturforsch Sect C J Biosci. 2011;66C(5–6):283–6.

74. Wang B, Pan X, Jia J, Xiong W, Manirafasha E, Ling X, Yinghua L. Strategy and regulatory mechanisms of glutamate feeding to enhance astaxan‑thin yield in Xanthophyllomyces dendrorhous. Enzyme Microb Technol. 2019;125:45–52.

75. Contreras G, Barahona S, Rojas MC, Baeza M, Cifuentes V, Alcaíno J. Increase in the astaxanthin synthase gene (crtS) dose by in vivo DNA fragment assembly in Xanthophyllomyces dendrorhous. BMC Biotechnol. 2013;13:1–10.

76. Barbachano‑Torres A, Castelblanco‑Matiz LM, Ramos‑Valdivia AC, Cerda‑García‑Rojas CM, Salgado LM, Flores‑Ortiz CM, Pnce‑Noyola T. Analysis of proteomic changes in colored mutants of Xanthophyllomyces dendrorhous (Phaffia rhodozyma). Arch Microbiol. 2014;196(6):411–21.

77. Retamales P, Hermosilla G, León R, Martínez C, Jiménez A, Cifuentes V. Development of the sexual reproductive cycle of Xanthophyllomyces dendrorhous. J Microbiol Methods. 2002;48(1):87–93.

78. Ang‑Feng S, Khaw SY, Few LL, See Too WC, Chew AL. Isolation of a stable astaxanthin‑hyperproducing mutant of Xanthophyllomyces dendrorhous through random mutagenesis. Appl Biochem Microbiol. 2019;55(3):255–63.

79. Gil‑Hwan AN, Schuman DB, Johnson EA. Isolation of Phaffia rhodozyma mutants with increased astaxanthin content. Appl Environ Microbiol. 1989;55(1):116–24.

80. Lewis MJ, Ragot N, Berlant MC, Miranda M. Selection of astaxanthin‑overproducing mutants of Phaffia rhodozyma with β‑ionone. Appl Environ Microbiol. 1990;56(9):2944–5.

81. Castelblanco‑Matiz LM, Barbachano‑Torres A, Ponce‑Noyola T, Ramos‑Valdivia AC, Cerda García‑Rojas CM, Flores‑Ortiz CM, Barahona‑Crisós‑tomo SK, Baeza‑Cancino ME, Alcaíno‑Gorman J, Cifuentes‑Guzmán VH. Carotenoid production and gene expression in an astaxanthin‑overpro‑ducing Xanthophyllomyces dendrorhous mutant strain. Arch Microbiol. 2015;197(10):1129–39.

82. Chun SB, Chin JE, Bai S, An GH. Strain improvement of Phaffia rho-dozyma by protoplast fusion. FEMS Microbiol Lett. 1992;93(3):221–6.

83. Farías‑Álvarez LJ, Gschaedler‑Mathis A, Sánchez‑Ortiz AF, Femat R. Xan-thophyllomyces dendrorhous physiological stages determination using combined measurements from dielectric and Raman spectroscopies, a cell counter system and fluorescence flow cytometry. Biochem Eng J. 2018;136:1–8.

84. Fontana GA, Hess D, Reinert JK, Mattarocci S, Falquet B, Klein D, Shore D, Thoma NH, Rass U. Rif1 S‑acylation mediates DNA double‑strand break repair at the inner nuclear membrane. Nat Commun. 2019;10(1):1–14.

85. Albertin W, Marullo P. Polyploidy in fungi: evolution after whole‑genome duplication. Proc R Soc B Biol Sci. 2012;279(1738):2497–509.

86. Alcaíno J, Barahona S, Carmona M, Lozano C, Marcoleta A, Niklitschek M, Sepúlvida D, Baeza M, Cifuentes V. Cloning of the cytochrome p450 reductase (crtR) gene and its involvement in the astaxanthin biosynthe‑sis of Xanthophyllomyces dendrorhous. BMC Microbiol. 2008;8:1–13.

87. Córdoba‑Castro NM, Acero‑ Reyes NL, Duque‑Buitrago LF, Jiménez‑Aguilar LJ, Serna‑Jiménez JA. Obtención y caracterización de astaxan‑tina de la microalga Haematococcus pluvialis. UGCiencia. 2015;21:73–82.

88. Yamamoto K, Hara KY, Morita T, Nishimura A, Ishii J, Ogino C, Kizaki N, Kondo A. Enhancement of astaxanthin production in Xanthophyl-lomyces dendrorhous by efficient method for the complete deletion of genes. Microb Cell Fact. 2016;15:155.

89. Girard P, Falconnier B, Bricout J, Vladescu B. β‑Carotene produc‑ing mutants of Phaffia rhodozyma. Appl Microbiol Biotechnol. 1994;41(2):183–91.

90. von Oppen‑Bezalel L, Fishbein D, Havas F, Ben‑Chitrit O, Khaiat A. The photoprotective effects of a food supplement tomato powder rich in phytoene and phytofluene, the colorless carotenoids, a preliminary study. Glob Dermatol. 2015;2(4):178–82.

91. An GH, Cho MH, Johnson EA. Monocyclic carotenoid biosynthetic path‑way in the yeast Phaffia rhodozyma (Xanthophyllomyces dendrorhous). J Biosci Bioeng. 1999;88(2):189–93.

92. Xie W, Lv X, Ye L, Zhou P, Yu H. Construction of lycopene‑overproducing Saccharomyces cerevisiae by combining directed evolution and meta‑bolic engineering. Metab Eng. 2015;30:69–78.

93. Yamada R, Yamauchi A, Ando Y, Kumata Y, Ogino H. Modulation of gene expression by cocktail Δ‑integration to improve carotenoid production in Saccharomyces cerevisiae. Bioresour Technol. 2018;268:616–21.

94. Córdova P, Gonzalez AM, Nelson DR, Gutiérrez MS, Baeza M, Cifuentes V, Alcaíno J. Characterization of the cytochrome P450 monooxygenase genes (P450ome) from the carotenogenic yeast Xanthophyllomyces dendrorhous. BMC Genom. 2017;18(1):1–13.

95. Loto I, Gutiérrez MS, Barahona S, Sepúlveda D, Martínez‑Moya P, Baeza M, Cifuentes V, Alcaíno J. Enhancement of carotenoid production by disrupting the C22‑sterol desaturase gene (CYP61) in Xanthophyllomy-ces dendrorhous. BMC Microbiol. 2012;12:235.

96. Zhou P, Xie W, Li A, Wang F, Yao Z, Bian Q, Zhu Y, Yu H, Ye L. Alleviation of metabolic bottleneck by combinatorial engineering enhanced asta‑xanthin synthesis in Saccharomyces cerevisiae. Enzyme Microb Technol. 2017;100:28–36.

97. Jiang G, Yang Z, Wang Y, Yao M, Chen Y, Xiao W, Yuan Y. Enhanced asta‑xanthin production in yeast via combined mutagenesis and evolution. Biochem Eng J. 2020;156:107519.

98. Wozniak A, Lozano C, Barahona S, Niklitschek M, Marcoleta A, Alcaíno J, Sepúlvida D, Baeza M, Cifuentes V. Differential carotenoid production and gene expression in Xanthophyllomyces dendrorhous grown in a non fermentable carbon source. FEMS Yeast Res. 2011;11(3):252–62.

99. Silva CM, de Borba TDM, Kalil SJ, de Burkert JFM. Raw glycerol and parboiled rice effluent for carotenoid production: effect of the com‑position of culture medium and initial pH. Food Technol Biotechnol. 2016;54(4):489–96.

100. Verwaal R, Wang J, Meijnen JP, Visser H, Sandmann G, Van Den Berg JA, van Ooyen JJ. High‑level production of beta‑carotene in Saccha-romyces cerevisiae by successive transformation with carotenogenic genes from Xanthophyllomyces dendrorhous. Appl Environ Microbiol. 2007;73(13):4342–50.

101. Gómez M, Gutiérrez MS, González AM, Gárate‑Castro C, Sepúlveda D, Barahona S, Baeza M, Cifuentes V, Alcaíno J. Metallopeptidase Stp1 activates the transcription factor Sre1 in the carotenogenic yeast Xan-thophyllomyces dendrorhous. J Lipid Res. 2020;61(2):229–43.

102. Mlíčková K, Roux E, Athenstaedt K, d’Andrea S, Daum G, Chardot T, Jean‑Marc N. Lipid accumulation, lipid body formation, and acyl coenzyme A oxidases of the yeast Yarrowia lipolytica. Appl Environ Microbiol. 2004;70(7):3918–24.

103. Hong J, Park SH, Kim S, Kim SW, Hahn JS. Efficient production of lycopene in Saccharomyces cerevisiae by enzyme engineering and increasing membrane flexibility and NAPDH production. Appl Microbiol Biotechnol. 2019;103(1):211–23.

104. Chen YT, Chao WC, Kuo HT, Shen JY, Chen IH, Yang HC, Wang JS, Lu JF, Cheng RP, Chou PT. Probing the polarity and water environment at the protein–peptide binding interface using tryptophan analogues. Biochem Biophys Rep. 2016;7:113–8.

105. Asadollahi MA, Maury J, Patil KR, Schalk M, Clark A, Nielsen J. Enhancing sesquiterpene production in Saccharomyces cerevisiae through in silico driven metabolic engineering. Metab Eng. 2009;11(6):328–34.

106. Pan X, Wang B, Gerken HG, Lu Y, Ling X. Proteomic analysis of astaxan‑thin biosynthesis in Xanthophyllomyces dendrorhous in response to low carbon levels. Bioprocess Biosyst Eng. 2017;40(7):1091–100.

Page 17 of 17Torres‑Haro et al. Microb Cell Fact (2021) 20:175

• fast, convenient online submission

•

thorough peer review by experienced researchers in your field

• rapid publication on acceptance

• support for research data, including large and complex data types

•

gold Open Access which fosters wider collaboration and increased citations

maximum visibility for your research: over 100M website views per year •

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions

Ready to submit your researchReady to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from:

107. Wang C, Long X, Mao X, Dong H, Xu L, Li Y. SigN is responsible for differentiation and stress responses based on comparative proteomic analyses of Streptomyces coelicolor wild‑type and sigN deletion strains. Microbiol Res. 2010;165(3):221–331.

108. Tahara EB, Barros MH, Oliveira GA, Netto LES, Kowaltowski AJ. Dihydroli‑poyl dehydrogenase as a source of reactive oxygen species inhibited by caloric restriction and involved in Saccharomyces cerevisiae aging. FASEB J. 2007;21(1):274–83.

109. Griffith CL, Klutts JS, Zhang L, Levery SB, Doering TL. UDP‑glucose dehy‑drogenase plays multiple roles in the biology of the pathogenic fungus Cryptococcus neoformans. J Biol Chem. 2004;279(49):51669–76.

110. Wong KH, Struhl K. The Cyc8–Tup1 complex inhibits transcription primarily by masking the activation domain of the recruiting protein. Genes Dev. 2011;25(23):2525–39.

111. Córdova P, Alcaíno J, Bravo N, Barahona S, Sepúlveda D, Fernández‑Lobato M, Baeza M, Cifuentes V. Regulation of carotenogenesis in the red yeast Xanthophyllomyces dendrorhous: the role of the transcriptional co‑repressor complex Cyc8–Tup1 involved in catabolic repression. Microb Cell Fact. 2016;15(1):1–19.

112. Carimi F, Zottini M, Formentin E, Terzi M, Lo SF. Cytokinins: new apop‑totic inducers in plants. Planta. 2003;216(3):413–21.

113. Cuellar‑Bermudez SP, Aguilar‑Hernández I, Cardenas‑Chavez DL, Ornelas‑Soto N, Romero‑Ogawa MA, Parra‑Saldivar R. Extraction and purification of high‑value metabolites from microalgae: essen‑tial lipids, astaxanthin and phycobiliproteins. Microb Biotechnol. 2015;8(2):190–209.

114. Weeks ME, Sinclair J, Butt A, Chung YL, Worthington JL, Wilkinson CRM, Griffiths J, Jones N, Waterfield MD. A parallel proteomic and metabolomic analysis of the hydrogen peroxide‑ and Sty1p‑depend‑ent stress response in Schizosaccharomyces pombe. Proteomics. 2006;6(9):2772–96.

115. Sauer U, Canonaco F, Heri S, Perrenoud A, Fischer E. The soluble and membrane‑bound transhydrogenases UdhA and PntAB have divergent functions in NADPH metabolism of Escherichia coli. J Biol Chem. 2004;279(8):6613–9.

116. Tsuruta H, Paddon CJ, Eng D, Lenihan JR, Horning T, Anthony LC, Regentin R, Keasling JD, Renninger N, Newman JD. High‑level produc‑tion of amorpha‑4, 11‑diene, a precursor of the antimalarial agent artemisinin, in Escherichia coli. PLoS ONE. 2009;4(2):e4489.

117. Chen H, Liu C, Li M, Zhang H, Xian M, Liu H. Directed evolution of meva‑lonate kinase in Escherichia coli by random mutagenesis for improved lycopene. RSC Adv. 2018;8(27):15021.

118. Choi HS, Lee SY, Kim TY, Woo HM. In silico identification of gene ampli‑fication targets for improvement of lycopene production. Appl Environ Microbiol. 2010;76(10):3097–105.

119. Gutiérrez MS, Campusano S, González AM, Gómez M, Barahona S, Sepúlveda D, Espenshade PJ, Fernández‑Lobato M, Baeza M, Cifuentes V, Alcaino J. Sterol regulatory element‑binding protein (Sre1) promotes the synthesis of carotenoids and sterols in Xanthophyllomyces den-drorhous. Front Microbiol. 2019;10(MAR):1–16.

120. Hughes AL, Todd BL, Espenshade PJ. SREBP pathway responds to sterols and functions as an oxygen sensor in fission yeast. Cell. 2005;120(6):831–42.

121. Bien CM, Espenshade PJ. Sterol regulatory element binding proteins in fungi: hypoxic transcription factors linked to pathogenesis. Eukaryot cell. 2010;9(3):352–9.

122. Jakočiūnas T, Pedersen LE, Lis AV, Jensen MK, Keasling JD. CasPER, a method for directed evolution in genomic contexts using mutagenesis and CRISPR/Cas9. Metab Eng. 2018;48:288–96.

123. Ducrey‑Santopietro LM, Spencer JFT, Spencer DM, Siñeriz F. Effects of oxidative stress on the production of carotenoid pigments by Phaffia rhodozyma (Xanthophyllomyces dendrorhous). Folia Microbiol. 1998;43(2):173–6.

124. Domonkos I, Kis M, Gombos Z, Ughy B. Carotenoids, versatile compo‑nents of oxygenic photosynthesis. Prog Lipid Res. 2013;52:539–61.

125. Arunkumar R, Gorusupudi A, Bernstein PS. The macular carotenoids: a biochemical overview. Biochim Biophys Acta Mol Cell Biol Lipids. 2020;1865:158617.

126. National Center for Biotechnology Information. https:// www. ncbi. nlm. nih. gov/. Accessed Feb 2021.

127. Ferreira P, Carro J, Serrano A, Martínez AT. A survey of genes encoding H2O2‑producing GMC oxidoreductases in 10 polyporales genomes. Mycologia. 2015;107(6):1105–19.

128. Sützl L, Foley G, Gillam EMJ, Bodén M, Haltrich D. The GMC superfam‑ily of oxidoreductases revisited: analysis and evolution of fungal GMC oxidoreductases. Biotechnol Biofuels. 2019;12(1):1–18.

129. EMBL‑EBI. EnsemblFungi. https:// fungi. ensem bl. org/ index. html. Accessed Jan 2020.

130. UniProt. Subcellular location. https:// www. unipr ot. org/. Accessed June 2021.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims in pub‑lished maps and institutional affiliations.